New Delhi: Nvidia has announced a new space-centric computing platform that will take artificial intelligence into space, a significant move toward space-based
data centres. This was announced during the company’s GTC 2026 conference when CEO Jensen Huang said that space computing was the next frontier of AI as more and more data is generated outside of Earth.
This new system is named the Vera Rubin Space-1 module and is designed to be used in extreme space conditions when size, weight and power are closely constrained. According to Nvidia, it will allow real-time data processing, autonomous decision-making, and advanced analytics to be performed on board, without relying on ground-based infrastructure.
Built for harsh space conditions
The Space-1 module is a packaged combination of Nvidia’s IGX Thor and Jetson Orin to provide high-performance AI computing in small, energy-efficient packages. The chips are designed with space missions in mind, and the conventional cooling does not achieve the desired cooling since convection is not present. Rather, Nvidia is coming up with new procedures to handle heat using radiation.
The company claims that its Rubin GPU can achieve much better AI performance on space-based inference than previous systems can achieve. This allows satellites and orbital platforms to handle gigantic amounts of data in real-time to enhance responsiveness and lessen delays associated with sending data back to Earth.
Powering orbital data centers
The recent thrust of Nvidia is in line with the increasing concerns with orbital data centres, which are supposed to serve the increasing energy demands on the Earth. Using solar power in space in large quantities, the systems may provide a scalable alternative to the traditional data centres, which are becoming more and more accused of consuming electricity.
However, challenges remain. Initial expenses, scarcity of rockets and technical challenges, including cooling and lifespan, remain obstacles to adopting them. Despite this, companies are pursuing the idea as the AI workloads keep on growing at a high rate.
Strong industry partnerships
Some space and satellite firms, such as Axiom Space, Planet Labs, Starcloud and Kepler Communications, already collaborate with Nvidia to implement its technology into spacecraft of the future. Such alliances are meant to inform smarter satellites that are able to process data onboard rather than depending on ground stations much.
Applications that are likely to be supported by the technology include geospatial intelligence, environmental surveillance and autonomous spacecraft missions. Nvidia expressed its larger ecosystem, consisting of ground-based GPUs such as the RTX PRO 6000 Blackwell, will be used to complement space-based systems for large-scale data processing.
The shift is a larger industry trend in the expansion of cloud and AI infrastructure to space. As technology firms compete to create next-generation personal computers, orbital platforms are soon projected to be one of the most important layers of international digital infrastructure.
Although the Vera Rubin Space-1 module should become marketable at a later time, Nvidia IGX Thor and Jetson Orin have already been released, so there is a strong indication that the age of AI-driven space operations is not too far off.
